Supersymmetric Hidden Sectors for Heterotic Standard Models

Supersymmetric Hidden Sectors for
Heterotic Standard Models

VoLKER Braun^, Yang-Hui He^, and Burt A. Ovrut^

^Dublin Institute for Advanced Studies
10 Burlington Road
Dublin 4, Ireland

^Department of Mathematics, City University
London, EClV OHB, U.K.

School of Physics, NanKai University
Tianjin, 300071, P.R. China
Merton College, University of Oxford, 0X1 4JD, U.K.

^Department of Physics

University of Pennsylvania
Philadelphia, PA 19104-6396

Abstract

Within the context of the weakly coupled Eg x Es heterotic string, we
study the hidden sector of heterotic standard model compactifications to four-
dimensions. Specifically, we present a class of hidden sector vector bundles —
composed of the direct sum of line bundles only — that, together with an ef-
fective bulk five-brane, renders the heterotic standard model entirely N = 1
supersymmetric. Two explicit hidden sectors are constructed and analyzed in
this context; one with the gauge group E^ x U{1) arising from a single line
bundle and a second with an SO (12) x ^(1) x gauge group constructed
from the direct sum of two line bundles. Each hidden sector bundle is shown
to satisfy all requisite physical constraints within a finite region of the Kahler
cone. We also clarify that the first Chern class of the line bundles need not be
even in our context, as has often been imposed in the model building literature.

Email: vbraunOstp . dias . ie, yang-hui . heOmerton . ox . ac . uk, ovrutQelcapitan . hep . upenn . edu

Contents

1 Introduction [T]

2 The Compactification Vacuum [3]

2.1 The Calabi-Yau Threefold |3]

2.2 The Observable Sector Gauge Bundle

2.3 The Hidden Sector Gauge Bundle IH]

2.4 Wrapped Five-Branes El

3 The Vacuum Constraint Conditions [8]

3.1 Anomaly Cancellation |8]

3.2 Slope-Stability of the Observable Sector Bundle |9]

3.3 = 1 Supersymmetric Hidden Sector Bundle [10]

3.4 Gauge Threshold Corrections [12]

4 Specific Examples 1131

4.1 Constraints for a Single Line Bundle [I3]

4.2 An X U{1) Hidden Sector [15]

4.3 Hidden Sector Matter Spectrum [16]

4.4 Constraints for Two Line Bundles [20]

4.5 An 50(12) x U(l) x U{1) Hidden Sector [SD

Appendix A Line Bundles and Eg [23]

A.l Induced Bundles [H

A. 2 Consistency Conditions [25]

A. 3 Chern Classes and the Anomaly [25]

A.4 Example [26]

Bibliography [27]

1 Introduction

The ten-dimensional theory of the massless modes of weakly coupled x E^ heterotic
theory can arise in two ways. The first is directly from the E'g x i^g heterotic superstring
after decoupling the heavy string modes [l||2]. The second follows from compactifying
eleven-dimensional M-theory on an /'L2 orbifold in the limit of small radius [3]-[9].
Either way, the ten-dimensional effective action is a A^ = 1 supersymmetric theory
with a metric, dilaton, two-form and E^ x E^ gauge fields as the bosonic components.
In addition, it can contain topological five-branes. This can be reduced to an A^ =
1 supersymmetric theory in four dimensions by appropriately compactifying on a

1

smooth Calabi-Yau threefold supporting gauge fields satisfying the hermitian Yang-
Mills equations 10,11 with the five-branes wrapped on holomorphic curves

12

The

choice of the Calabi-Yau manifold and the gauge field background, that is, a slope-
stable holomorphic vector bundle with vanishing slope, determines the low energy

gauge group, spectrum, and coupling parameters 13[jl7
In U

I , it was shown that the low energy spectrum of a specific Calabi-Yau three-
fold and holomorphic vector bundle with structure group SU{4:) C Es is exactly that
of the minimal supersymmetric standard model (MSSM) with three right-handed neu-
trinos and one pair of Higgs-Higgs conjugate chiral multiplets. There are no exotic
or vector-like pairs of superfields. It was demonstrated in that over a specific
subspace of the Kahler cone, this vector bundle is slope-stable with vanishing slope
and, hence, the low energy theory is = 1 supersymmetric. For these reasons, this
vacuum of the observable Eg sector of the theory was called the heterotic standard
model. To complete the vacuum, it is essential to present the explicit vector bundle for
the second Eg hidden sector. In previous work 20-22 , it was expedient to choose this

bundle to be trivial, requiring a five-brane sector with non-effective cohomology class
to satisfy the anomaly constraint. This anti-brane allows one to raise the potential
energy of the vacuum from negative to a small, positive cosmological constant [23] ,

that is, the heterotic equivalent of the KKLT mechanism [24]. It was shown in 19
that, for vanishing five-brane class, a hidden sector SU{n) bundle satisfying the Bo-
gomolov bound [25j could exist in a subspace intersecting the region of stability of
the observable sector vector bundle. The Bogomolov bound is a necessary condition
for the hidden bundle to be = 1 supersymmetric. At least on certain manifolds,
satisfying this bound is also sufficient 26 28 . However, an explicit example was never
constructed.

In a series of papers, both in weakly coupled 29 33 and strongly coupled [7|[8 ,34
Eg X Eg heterotic theory, the first non-trivial corrections — string one-loop and M-
theory order k^^^ respectively — beyond tree level were constructed. These were pre-
sented for both the Fayet-Iliopoulos (FI) terms associated with anomalous U{1) gauge
factors and for the gauge threshold corrections. Furthermore, it was emphasized
in [29 -33 that gauge bundles with U{n) structure groups, in addition to SU{n) bun-
dles, should be more closely scrutinized. Similarly, the appearance of line bundles on
the stability wall boundaries of SU (n) bundles was recognized in 34 , 35 , and applied
more widely to construct new realistic models in 36 38 . In this paper, within the

context of the weakly coupled Es x Eg heterotic string, we will apply these ideas to
construct a completely A^ = 1 supersymmetric hidden sector for the heterotic stan-
dard model. This hidden sector will have a non-trivial, but effective, five-brane class
and a hidden sector vector bundle composed of a direct sum of line bundles. Together
with the observable sector SU{4) vector bundle, this provides an explicit A^ = 1
compactification vacuum for the heterotic standard model.

Specifically, we will do the following. In Section 2 we review the Calabi-Yau three-

2

fold and observable sector gauge bundle of the heterotic standard model introduced
The generic form of the hidden sector bundle as a sum of line bundles is

18,19

m

presented, as is an analysis of the explicit embedding of a line bundle into an Eg vector
bundle. Five-branes are also briefly discussed. [Section 3] consists of an analysis of the
four constraint conditions required for our compactification vacuum; namely, anomaly
cancellation, slope-stability of the observable sector bundle, = 1 supersymmetry of
the hidden sector bundle and positivity of the threshold corrected gauge parameters.
The explicit constraint equations for each of these conditions are given. Two specific
examples are then presented in Section 4 The first is obtained using a single line bun-
dle as the hidden sector bundle. Its exact embedding into the hidden Eg is discussed,
and it is shown to satisfy all required constraint conditions in a region of the Kahler
cone. The associated low energy gauge group and spectrum of the hidden sector is
presented. As a second example, we choose the hidden sector bundle to be a sum of
two line bundles. Again, we elucidate its exact embedding into the hidden Es, show
that it satisfies all necessary constraints in a region of the Kahler cone and discuss
the associated low energy gauge group. Finally, important technical issues regarding

the embedding of line bundles into an Eg vector bundle are discussed in [Appendix A

The analysis in this paper has a direct, but non-trivial, extension to strongly cou-

This will be presented in a future publication.

pled heterotic M-theory 3 4 8 9 39 44

2 The Compactification Vacuum

A four-dimensional, A^ = 1 super symmetric effective theory of the weakly coupled
Eg X Es heterotic string is obtained as follows. First, one compactifies ten-dimensional
spacetime on an appropriate Calabi-Yau threefold, X. Second, it is necessary to con-
struct two Eg gauge bundles over the Calabi-Yau manifold. Two popular methods are
to utilize a slope-stable holomorphic vector bundle with vanishing first Chern class
or a line bundle. The latter is automatically slope-stable, but imposes the additional
physical constraint that the Fl-term must vanish. The associated gauge symme-
tries, spectrum and coupling parameters of the four- dimensional theory depend on
the specific choice of the compactification. In this paper, we will examine a physi-
cally relevant subset of such vacua; namely, those with the Calabi-Yau threefold and

observable sector vector bundle of the heterotic standard model 18 , 19 .

2.1 The Calabi-Yau Threefold

The Calabi-Yau manifold X is chosen to be a torus-fibered threefold with fundamental
group 7Ti{X) = Z3 X Z3. Specifically, it is a fiber product of two rational elliptic dFg

surfaces, that is, a self-mirror Schoen threefold 43 ,45
a freely acting Z3 x Z3 isometry. Its Hodge data is h
are three Kahler and three complex structure moduli. The complex structure moduli

49 quotiented with respect to
= /i^'^ = 3 and, hence, there

3

will play no role in the present paper and we will ignore them. Relevant here is the
degree-two Dolbeault cohomology group

i/^'^(X,C) = spanc{wi, W2, ^3}, (1)

where Ui = u^^idz'^dz^ are dimensionless harmonic (1, l)-forms on X with the property

cus A cjs = 0, Wi A

Defining the intersection numbers as

1

3u;i A Wi, UJ2 /\oj'i = 3a;2 A uj2-

d.

ijk

LOi A tUj A ojk, i, j, k = 1,2,3

(3)

where w is a reference volume of dimension (length)^, it follows from ^ that

/(0,|,0) (0,1,0)\

(1,0,0) (1,0,0) .
\(0,1,0) (1,0,0) (0,0,0)/

d.

ijk

(4)

The {ij'}-th entry in the matrix corresponds to the triplet {d^ijjk\k = 1, 2, 3).

Our analysis will require the Chern classes of the tangent bundle TX. Noting that
the associated structure group is SU{3) C SO (6), it follows that rank(TX) = 3 and
ci(TX) = 0. Furthermore, the self-mirror property of this specific threefold implies
C3(TX) = 0. Finally, we find that

C2(TX)

,2/3

- (I2wi A Wi + 12^2 A ^2) .

(5)

We will use the fact that if one chooses the generators of SU (3) to be hermitian, then
the second Chern class of the tangent bundle can be written as

C2(TX)

167r2

tiRAR,

(6)

where R is the Lie algebra valued curvature two-form.

Note that H^'° = H^'"^ = on a Calabi-Yau threefold. It follows that H^'^{X, C) =
H'^{X, M) and, hence, Ui, i = 1,2,3 span the real vector space H'^{X, M). Furthermore,
it was shown in 50 that the curve Poincare dual to each two-form Ui is effective.
Therefore, the Kahler cone is the positive octant

/C = Hl{X, R) C H'^iX, R). (7)

The Kahler form, defined to be = igaS where gai is the Calabi-Yau metric, can be
any element of /C. That is, the Kahler form can be expanded as

a uji

d > 0,

1,2,3.

(8)

4

The real, positive coefficients a* are the three (1, 1) Kahler moduh of the Calabi-Yau
threefold. Here, and through this paper, upper and lower H^'^ indices are summed
unless otherwise stated. The dimensionless volume modulus is defined by

V = - [ V^g (9)

and, hence, the dimensionful Calabi-Yau volume is vV. Using the definition of the
Kahler form and ([s]), V can be written as

If 1 • •

V = — / cu Acu Aco = -dijka'a^a''. (10)

6v Jx 6

It is useful to express the three (1,1) moduli in terms of V and two additional inde-
pendent moduli. This can be accomplished by defining the scaled shape moduli

b' = V-^/^a\ i = l,2,3. (11)

It follows from ( 10 ) that they satisfy the constraint

dijuVVh^ = 6 (12)

and, hence, represent only two degrees of freedom. Finally, note that all moduli defined
thus far, that is, a\ V and b\ are functions of the four coordinates x^, /x = 0, . . . ,3
of Minkowski space M4.

2.2 The Observable Sector Gauge Bundle

The Es X Es vector bundle V over X is a direct sum of an observable sector bundle,
V^^\ whose structure group is embedded in the ffist Eg factor, with a hidden sector
bundle, V^'^\ with structure group in the second E^. V^^^ is chosen to be holomorphic
with structure group S'f/(4) C Eg, thus breaking

Es — > Spin{10). (13)

Our analysis will require the Chern classes of V^^^^ Since the structure group is S'f/(4),
it follows immediately that ran k(V(i)) = 4 and Ci(y(^)) = 0. The heterotic standard
model is constructed to have the observed three chiral families of quarks/leptons and,
hence, V^*-^^ must be chosen so that c^iV^^^) = 3. Finally, we find that

C2{V^^^) = (wi A tJi + 4 ^2 A ^2 + 4 wi A ws) . (14)

Here, and below, it will be useful to note the following. Let V be an arbitrary vector
bundle with structure group G, and J-"^ the associated Lie algebra valued two-form
gauge field strength. If the generators of G are chosen to be hermitian, then

trc^^ A = ch2(V) = ^Ci(V) A Ci(V) - C2(V), (15)

5

where ch2(V) is the second Chern character of V. Furthermore, we denote by tr^ the
trace in the fundamental representation of the structure group G of the bundle. When
applied to the vector bundle V^^^ in the observable sector, it follows from ciiV^^'^) =
that

C2iV^'^) = -^,tTsui4) F^'^ A = tr^, F^'^ A F^'\ (16)

where F^^^ is the gauge field strength for the visible sector bundle V^^^ and tr^;^
indicates the trace is over the fundamental 248 representation of Eg. Note that the
conventional normalization of the trace tr^;^ includes a factor of t|j, the inverse of the
dual Coxeter number of E^. We have expressed C2{V^^^) in terms of tr^;^ since the
fundamental SU{4:) representation must be embedded into the adjoint representation
of in the observable sector.

To preserve = 1 supersymmmetry in four- dimensions, V^'-^-' must be both slope-

stable and have vanishing slope 51 ,52 . In the context of this paper, these constraints
are most easily examined in the (1 = 4 effective theory and, hence, will be discussed
below. Finally, when two flat Wilson lines are turned on, each generating a different
Z3 factor of the Z3 x Z3 holonomy of X, the observable gauge group is further broken
to

Spin{10) SU{3)c X SU{2)l x f/(l)y x U{1)b-l. (17)
2.3 The Hidden Sector Gauge Bundle

In the hidden sector, the vector bundle V^'^^ introduced in this paper will be con-
structed entirely as the sum of holomorphic line bundles. Let us briefly review the
properties of such bundles on our specific geometry. Line bundles are classified by the
divisors of X and, hence, equivalently by the elements of the integral cohomology

H'^{X, Z) = {auJi + bL02 + cuj3 \ a,b,ceZ, a + b = mod 3}. (18)

It is conventional to denote the line bundle associated with the element aui + buj2 + cu3
of if2(x,Z) as

Ox(a,6,c). (19)

Note that the ui, U2, are the natural basis of invariant integral forms on the covering
space. In order to correspond to integral forms on the quotient Calabi-Yau manifold
X, an element aui + bu2 + cu3 has to satisfy the additional constraint a + b = mod 3
in order to be integral. This can also be seen from the intersection numbers (|4]), which
are naively fractional. Only the intersection of classes satisfying a + b = mod 3 is
integral. For the purposes of constructing a heterotic gauge bundle from a line bundle
Ox (a, b, c), this is the only constraint required on the integers a, b, c. In particular, as
explained in Appendix A[ it is not necessary to impose that they be even for there to

exist a spin structure on V^'^\ Although the auxiliary line bundle is not spin if a, b,
c are not even, the Es bundle is always spin.

6

We will choose the the hidden sector bundle to be

R

r=l

where

mod 3,

R

(20)

(21)

for some positive integer R. The structure group is U{1)^, where each U{1) factor
has a specific embedding into the hidden sector Eg gauge group. It follows from the
definition that rank(y*^^^) = R and that the first Chern class is

R

ci(l^(2)) = J]ci(L,), ci(L,,)

r=l

(22)

Note that since is a sum of holomorphic line bundles, C2{V^'^^) = c^iV^"^^) = 0.
However, the relevant quantity for the hidden sector vacuum is related to the second
Chern character given in (15). Defining F^"^^ to be the gauge field strength for the
hidden sector bundle V^'^\ this becomes

R Ft

tr^(i). A = ch2(\/(2)) = J2 ^MLr) = ^ci(L,) A ci(L,) (23)

r=l

r=l

since C2{Lr) = 0. As in the observable sector, the U{1)^ fundamental representation
must be embedded into the adjoint representation of the hidden sector Eg. Hence, the
physically relevant quantity is proportional to tr^g of F^'^^ AF^'^\ where we remind the
reader that our normalization of the Eg trace includes the ^ as in (16). Specifically,

the term of interest is
1

R

2a

167r2

with a group-theoretic factor

tr^3 A = E ^ ^'^W ^r'^ ^ ^r'^ = E

r=l

r=l

tr^;^ Q

(24)

(25)

where Qr is the generator of the r-th U{1) factor embedded into the 248 representation
of the hidden sector Eg. Note that we have used (23) in going from the second to the
third term in (24).

The definition of with the coefficient | is, of course, a convention. However, it
is justified by the following computation which we leave as an exercise for the reader
to verify. Using the embedding U{1) C SU{2) C Ej defined by eqns. (84) and (52)
below, the normalized trace is given by tr^;^

30

■ 60 ■ 2. Therefore, the minimal

7

U{1) embedding in Eg leads to a^i^ = 1, explaining the conventional normalization
factor of I in (25). In fact, by comparing the usual formula for the Chern character of
a line bundle (23) with the Eg characteristic class (24), one might have guessed that
ttr is half-integral. However, this is not true and is always integral. This is also
crucially important for the contribution to the heterotic anomaly, which must be an
integral cohomology class, to be well-defined. In Appendix A, we will discuss this in
more detail.

2.4 Wrapped Five-Branes

In addition to the holomorphic vector bundles in the observable and hidden sectors,
the compactification can also contain five-branes wrapped on two-cycles n =
1, . . . , M in X. Cohomologically, each such five-brane is described by the (2,2)-form
Poincare dual to Cg"'', which we denote by W''"'\ Note that to preserve = 1
supersymmetry in the four-dimensional theory, these curves must be holomorphic
and, hence, each VT^") be an effective class.

3 The Vacuum Constraint Conditions

In order for the Calabi-Yau threefold X, the observable and hidden sector vector
bundles V^^\ V^*-^^ and the five-branes W^""^ discussed above to form a consistent com-
pactification, they must satisfy a set of physical constraints. These are the following.

3.1 Anomaly Cancellation

As discussed in [7[]9| [5l|[52] , anomaly cancellation requires that

11 1 "

- — tTRAR+ — tTE, A F(^) + — tr^, F^^) A F^^) - J] ly M = 0. (26)

571=1

Using ([6]), (16) and (23), (24) the anomaly cancellation condition can be expressed as

R

C2{TX) - C2(r(i)) + J2 aMLr) A Ci(L,) -W = 0, (27)
where W = ^^^^W^^'K

r=l

Condition (27) is expressed in terms of four-forms in H {X,M.). We find it easier

to analyze its consequences by writing it in the dual homology space H2{X,M.). In
this case, the coefficient of the i-th vector in the basis dual to (wi, ci;2, Ws) is given by

8

wedging each term in (27) with Ui and integrating over X. Using (|5|,(14) and the
intersection numbers (|3j),(|4j) gives

1

,1/3

X

C2(TX) -C2(K(i))) A a;i,2.3 = (|,i, -4).

(2J

Similarly, (|3|),(|4|) and (|22|) imply
1

,1/3

Ci{Lr) A Ci{Lr) Auji = dijkiW;^ ^ = 1, 2, 3.

Defining

,,1/3 / ^

(29)

(30)

it follows that the anomaly condition (27) can be expressed as

R

Wi={l I -4), + J] ard^.kiii'; > 0, ^ = 1, 2, 3.

(31)

r=l

The semi-positivity constraint on W follows from the requirement that it be an effec-
tive class to preserve = 1 supersymmetry.

3.2 Slope-Stability of the Observable Sector Bundle

As mentioned previously, to preserve = 1 supersymmmetry in four- dimensions the
holomorphic SU (4) vector bundle V^^'^ associated with the observable Eg gauge group
must be both

• slope-stable (to admit a solution to the Hermitian Yang-Mills equation), and

• have vanishing slope (because there is no FI term for SU{n) bundles).
Here, the slope of any bundle or sub-bundle J-" is defined as

1

/i(^)

rank(J-')f ^/^ Jx

Ci{J-') Alu A LO,

(32)

where u = a^Ui is the Kahler form as in ([s]). The rank-4 bundle V^^^ h as vani shing
slope since ci{V^^^) = 0. But, is it slope-stable? As shown in detail in
can identify a set of 7 "maximally destabilizing" line sub-bundles

19

23

, one

Ox(l,-l,-l), Ox(-l,l,-l), Ox(-2,2,0), Ox(2,-2,-l),
Ox(2,-5,l), Ox(l,-4,l), Ox(-4,l,l).

(33)

9

It is a sufficient condition for stability to have all of their slopes be negative simulta-
neously. This singles out the subspace of the Kahler cone defined by the following 7
inequalities

-3(a^ - a^)(a^ + + Qa^) - ISa^a^ <
3(a^ - a^)(a^ + + 6a^) - ISa^a"^ <
6{a^ - a^){a^ + a"^ + 6a^) <
-6(a^ - a^)(a^ + + 6a^) - ISa^a^ < (34)
-3(5a^ - 2a^){a^ + + 6a^) + 9a^a^ <
-3(4a^ - a^){a^ + + Qa^) + 9a^a^ <
3(a^ - 4a^)(a^ + + 6a^) + 9a^a^ < 0.

These can be slightly simplified into the statement that a\i = 1,2,3 must satisfy at
least one of the two sets of inequahties

a <a^<\%a and a-^ < — — — —\ or

2 ^q}' — 6(2^

(Vi" <2a' and 30^. _ j,,. < '''^ < ^r;^, )

The subspace satisfying (34) is a full-dimensional subcone of the Kahler cone /C
defined in Q. It is a cone because the inequalities are homogeneous. In other words,
only the angular part of the Kahler moduli are constrained but not the overall volume.
Hence, it is best displayed as a two-dimensional "star map" as seen by an observer at
the origin. This is shown in Figure 1 For Kahler moduli restricted to this subcone, the
four- dimensional low energy theory in the observable sector is = 1 supersymmetric.

3.3 AT = 1 Supersymmetric Hidden Sector Bundle

In the heterotic standard model vacuum, the observable sector vector bundle y*^^^ has
structure group SU{4). Hence, it does not lead to an anomalous U{1) gauge factor
in the observable sector of the low energy theory. However, the hidden sector bundle
V'^'^'^ introduced above consists entirely of a sum of line bundles and, therefore, has
structure group U{1)^. Each U{1) factor leads to an anomalous U{1) gauge group in
the four-dimensional effective field theory and, hence, an associated D-term.

Let Lr be any one of the sub-line bundles of V'^'^\ Then, it was shown in
the associated FI term is

oc /i(L,) - ^ y" ci(L,)A

/ R M \

^a,ci(L,) A Ci(L,) + -C2(TX) - ^(^ + X^fW^'^'^ \ , (36)

\s=l m=l /

10

Figure 1: Map projection of the unit sphere intersecting the Kdhler cone, that
is, the positive octant in M) ~ M^. The visible sector bundle

y^^^ is stable inside the red teardrop- shaped region /C*. Every point
in the projection represents a ray in the Kdhler cone. For example,
{a^,a'^,a^) = (0,1,0) generates the ray in the uj2 direction.

where ^i{Lr) is given in (32) and

9s

L = 27rVa'

(37)

are the string couphng and string length respectively. Furthermore, each A„ is a real
modulus that, together with a self-dual two- form B„, forms a tensor multiplet on
the six-dimensional worldvolume of the n-th five-brane. The normalization of these
moduli is chosen so that

^ [ 2' 2] ■

(3J

The first term on the right-hand side, that is, the slope of Lr defined in (32), is the
tree level result.

The remaining terms are the string one-loop string corrections first presented
in [29[ [32} [53] . The general form of each D-term is as the sum of 1) the moduli

dependent FI parameter (36) and 2) terms quadratic in the fields charged under the
gauge symmetry weighted by their specific charge. In this paper, for simplicity, we
will assume that all f/(l)^ charged zero-modes in the hidden sector have vanishing

11

expectation values. Then the hidden sector will be = 1 supersymmetric if and only
if the moduli-dependent FI parameter vanishes for each L^. That is,

ci(L,.) A a; A w - 5fg/g / Cl{Lr)^

X J X

/ R M \

5^a,ci(L,)Aci(L,) + -C2(TX)- J](i + Aj2iy("^) =o (39)

\s=l m=l /

for r = 1, . . . , i?. Using Q, Q, g, (g, (g, Q and noting from Q that

j^^C2{TX)Au, = {2,2,0)i, (40)

it follows that for each condition (39) can be written as

R

dijklia^a^

2/4

ail

M

,2/3

d^,kil. J2 + ^r(2, 2, 0), - 5^(1 + XmYiiW,^"^^ = (41)

s=l

m=l

where

V = -dijuda^a^.
6

(42)

3.4 Gauge Threshold Corrections

The gauge couplings of the non-anomalous components of the d = A gauge group, in
both the observable and hidden sectors, have been computed to the string one-loop

level in 29 33 . Including five-branes, these are given by

is''s

2v

Air

9

(1)2 6v

u Au Au- ^ / uA

X

X

\ m=l /

(43)

and

An

9

(2)2

X

u Au Au- / uA
2v

X

(R ^ Af ^

5^a,ci(L,) A ci(L,) + -C2(TX) - J^H + A™)'W^^"^
r=l m=l /

(44)

respectively. The first term on the right-hand side, that is, the volume V defined in

(10), is the tree level result. The remaining terms are the one-loop corrections first

presented in |29 .

12

Clearly, consistency of the d = 4 effective theory requires both g^^^'^ and g^^^'^ to
be positive. It follows that the moduli of the four-dimensional theory are constrained
to satisfy

uAuAcu-gllt / c^A ( - C2{V^'^)
■J -Ix J X

and

w A w A w - gll^ I u A

X

X

1 \

+ -C2(TX)-5^(i-Aj2W-("M >

m=l
. R

( QrCl (Lr) A Ci (Lr)

V r=l

1 \

m=l '

(45)

(46)

As in the previous subsections, one can use (|3]), (|4]), ([s]), (14), (29), (30) and (40) to
rewrite these expressions as

2/4 /

,,fca^a^a-^ - ( -(fa^ + \o? + 40^) + 2(ai + a^) - ^(1 - A^)2a^ W^^™) ) >

m=l

(47)

and

d,,fca'a^a^ - ( ci,,fca* ^ a.F/,' + 2(0^ + a') - ^^(^ + A^)'a* W^"^^ \ >

M

r=l

m=l

(4J

respectively.

4 Specific Examples

4.1 Constraints for a Single Line Bundle

We now present an explicit = 1 supersymmetric hidden sector for the weakly
coupled heterotic standard model that satisfies all vacuum constraints. To do this,
one must specify the number of line bundles Lr and their exact embeddings into
the hidden vector bundle Eg. Later in this section, we will consider the case of two

13

independent line bundles. However, for now we restrict ourselves to the simplest
example consisting only of a single line bundle

\/(2) = L, L = Ox{i\f,f) (49)

parametrized by integers

ni „2 ^3 ^ £i + £2 _ mod 3. (50)

5 ^ 1

Furthermore, the explicit embedding of L into is chosen as follows. First, recall
that

SU{2) xEtCEs (51)

is a maximal subgroup. With respect to SU{2) x Ej, the 248 representation of Eg
decomposes as

248 (1 , 133) © (2, 56) © (3, 1) . (52)

We embed the generator Q of the U(l) structure group of L — more specifically, the
generator of the S{U{1) x U{1)) Abelian group of the induced rank two bundle in
SU{2) — so that under SU{2) — > U{1) the two-dimensional SU{2) representation de-
composes as

2^10-1. (53)

It follows that under f/(l) x Ej

248 ^ (0, 133) © ((1, 56) © (-1, M)) © ((2, 1) © (0, 1) © (-2, 1)) . (54)

The generator Q of this embedding of the line bundle can be read off from expression
(54). Inserting this into (25), we find that

a = 1. (55)

Having presented the hidden sector vector bundle, one must specify the number
of five-branes. For simplicity, we assume that there is only one five-brane in this
example. It then follows from (31), (41), (47) and (48) that the constraints for this

14

explicit example are given by

= (|, |, -4)^ + > 0, i = 1, 2, 3 (56a)

dijkta^a'^

^2/3

-(^ + A)'m) =0,

dijka'a^a^ - - (|a^ + + 4a^) +

+2{a^ + a^) - (l - X)\'Wi^ > 0,

+2{a^ + a^)- {l + Xfa'W^ > 0.

(56b)

(56c)

(56d)

These constraints have to be solved simultaneously with the condition (35) for the
slope-stability of the observable E'g non-Abelian vector bundle. That is,

< < \l^a} and <

-(ai)2-3aV + (a2)2

or

5 1 2 o 1 A 2(a')' - 5(ai) 3

< < 2a^ and ^ — < a'^ <

2 30ai - 12a2

6a^ — 6a2

^^^1^2 _(al)2_3^1^2^/^2N2

6a^ — 6a2

(56e)

We now seek simultaneous solutions to eqns. (56a) to (56e)

4.2 An Ey X U{1) Hidden Sector

The first observation is that the system of equations (56b) to (56e) is homogeneous
with respect to the rescalinaM

ys'-s

2/4

1 2 3 ys<'s \ . , I ,.„1 ,.„2 ,.„3 ,.2ds''s

Uj ^ (Jj ^ (Jj

^2/3 j ^ \^^^ , na , fia , fJ, ^2/3
Therefore, one can absorb the coupling constants into the Kahler moduli a\ In other

/i > 0.

(57)

words, for a single U{1) the FI = equation (56b) fixes one Kahler modulus to a

certain numerical value measured in multiples of (%7l)^''^- This is tantamount to

setting ^fyi to unity, which we will do henceforth for simplicity.
Next, let us concentrate on the particular hidden line bundle

\/(2) = L = Ox(l,2,3),

(58)

^Note that the coupling constants appear only in the combination ^yf , which is a positive number.

15

that is,

(£^£^£3) = (1,2,3).

This choice cancels the anomaly with the effective five-brane curve class

W = (16,10,0) > 0.

(59)

(60)

Having fixed the line bundle, we proceed to solve the Fayet-Iliopoulos equation (56b])
for a^, giving us

-2{a^y - (a2)2 - 2Aa^a^ - lOSA^ - 108A + 117
6 (2ai + a2) '

(61)

This can then be substituted into equations ( 56c[ ), (56d) and (56e) to obtain a system
of polynomial inequalities in a^, and A.

Finally, one can scan through the range — ^ < A < | and plot the region of
validity in the a^-a^ plane to find solutions. For example, with the numerical value of
A = 0.496, which is close to the hidden wall, we do indeed find solutions which satisfy

all of the constraints. The region where all physical constraints equations (56c) to

(56e) are satisfied simultaneously is shown in yellow in Figure 2 For clarity, and to

make contact with the visible sector stability region plotted in Figure 1, we present
the same data in Figure 3 as a subset of the Kahler cone — that is, the positive octant

a , a , a > 0. This gives a multi-dimensional visualization of the facts that:

• The visible sector stability region (red) is a three-dimensional sub-cone of the
Kahler cone.

• The Fl-term stabilizes one particular combination of angular and radial Kahler
moduli, which is why the hidden sector U{1) constrains us to a two-dimensional
surface (blue) in the Kahler cone.

4.3 Hidden Sector Matter Spectrum

We can now compute the low energy EixU{l) particle spectrum from the cohomology

of the line bundle L. From the breaking pattern in (52) and (53), one can read off
the representation content. The spectrum is then determined by the cohomology
groups of the corresponding tensor power of the hne bundle. The even (odd) degree
cohomology groups count the left-chiral (right-chiral) supermultiplets. The result is
tabulated in the middle column of Table 1[ Here, we used the fact that the trivial
line bundle Ox on the threefold X has cohomology groups of dimension

h^{X, Ox) = h\X, Ox) = l

(62)

and is otherwise by the Calabi-Yau property of X. Moreover, L* is the dual bundle
to L and is the second tensor power of L.

16

0.7 0.8 0.9 1.0 1.1 1.2 1.3

Figure 2: The two-dimensional slice through the Kdhler cone where the FI-
term of the hidden line bundle L = Ox(l, 2, 3) with five-brane posi-
tion X = 0.496 vanishes. The slice is parametrized by (a"^,a^) with
given by ( |61[ ). In red, the visible sector stability condition, see
sub-figures a) and c). In blue, the region where the both the visible
and hidden sector gauge couplings are positive, see sub-figures b) and
c). Their intersection is drawn in yellow, see sub-figure c).

17

Figure 3: The Kdhler cone, in 3 dimensions (top) and the projection in radial directions
(bottom). The blue region /C^ is our hidden sector solution for L = Ox(l,2,3)
at X = 0.496. It shows the Kdhler moduli uj = a^wi + a^a;2 +a^W3 simultaneously
satisfying the FI = condition and the positivity of the visible and hidden sector
gauge couplings. The red region fC^ is the stability region of the visible sector

bundle from Figure 1. The intersection IC^ n JC^ is where all physical constraints
are satisfied.

18

U{1) X
representation

Multiplicity

Multiplicity
if L is ample

(0, 133)

h%X, Ox)

(1,56)

h\X,L) + h\X,L)

h\X,L)

(-1,56)

h%X, L*) + h\X, L*)

(2,1)

h'^{X,L^) + h\X, L^)

h\X,L^)

(-2,1)

h\X, L^*) + h\X, L^*)

(0,1)

h%X,Ox)

h\X, Ox)

Table 1: Matter spectrum for the — )• U{1) x Ej breaking pattern with a line
bundle L. The multiplicity counts the number of left-chiral N = 1
multiplets with the given gauge charge.

We therefore need to compute the dimensions of the cohomology groups H^{X, L),
H^{X,L*), H^{X,L^), and H\X,L^*). Because our line bundle L = 0^(1,2,3) has
only positive entries, it is ample. An immediate consequence is that

h\X, L) = h\X, L^) = 0, i^Q (63)

by the Kodaira vanishing theorem. Therefore, only h^{X,L) and /i°(X, L^) can be
non-zero. Furthermore, by Serre duality

h\X,L) = h^-\X,L*), h\X,L^) = e-\X,L^*), i = 0,1,2,3. (64)

It then follows that

h\X,L*) = h\X,L^*) = 0. (65)

The consequences of an ample line bundle L are also tabulated in the rightmost column
of ITable II

The dimensions of the two remaining non- vanishing cohomology groups, h^{X, L)
and h^{X,L'^), are then easily determined from the Atiyah-Singer index theorem. It
states that, for any vector bundle \^ on a Calabi-Yau threefold X,

3

xiV) = y2hl)'h'{X, V) = / ch(l^) A td(TX). (66)

In our present case of an ample line bundle ©^(^i, ^2, ^3), using (63) and (64), this
simplifies to

h\X,Ox{h,i2j3)) = ^ (^Ci(\/(2)) A C2(TX) + ch3(\/(^))

l"! 1 . 1 ' ... ^^^^

r,s,t=l

19

Hence, we find

/i°(X,L) = 8 for L = Ox(l,2,3) (68)

and, similarly, that

/i°(X,L2) = 58 for = Ox(2,4,6). (69)
In summary, the complete U{1) x Ej hidden sector matter spectrum of our model is

1 X (0,133) + 8 X (1,56) + 58 x (2,1) + 1 x (0,1). (70)
4.4 Constraints for Two Line Bundles

In the previous section, we discussed the case of the hidden sector being a single
line bundle and presented a detailed solution. One can easily move on to bundles of
higher rank and show that indeed there is a plenitude of solutions. This is clearly
desirable for model building since each independent U{1) imposes one Fayet-Iliopoulos
vanishing equation and, therefore, stabilizes one Kahler modulus. In this section, we
will consider the next simplest hidden sector consisting of the direct sum of two line
bundles

V^''^ = L,®L2, L^ = Ox{llllll). L2 = Ox{llAA) (71)

where

Z,^ + = mod 3, r = l,2. (72)

Furthermore, Li © L2 will be given the simplest simplest embedding into E^, namely
via

U{1) X U{1) X 50(12) c SU{2) X SU{2) x 50(12) c E^. (73)
The branching rules easily follow from those of the SU{2) x 50(12) maximal sub-

group of Ey in conjunction with (52). In particular, the adjoint representation of Eg

decomposes under SU{2) x SU{2) x 50(12) as

248 (1, 3, 1) © (3, 1, 1) © (2, 1, 32) © (1, 2, 32)
© (2, 2, 12) © (2, 2, 12) © (1, 1, 66).

One now has to choose the embedding of the generator Qr of each of the two
U{1) structure groups into the corresponding 5f/(2). We again pick the simplest
embedding, identifying the structure group of Li with the center of the first 5f/(2),

as in (53), and similarly for L2. Consequently, under U{1) x U{1) x 50(12) we have

the branching rule

248 (0, 2, 1) © (2, 0, 1) © (0, -2, 1) © (-2, 0, 1) © 2 x (0, 0, 1)

© (1,0,32) © (-1,0,32) © (0,1,32) © (0,-1,32) © (1,1,12) (75)
© (1, -1, 12) © (-1, 1, 12) © (-1, -1, 12) © (0,0,66).

20

As before, one can read off from (25) the group-theoretic embedding coefficients ai
and a2- They are given by

ai = 02 = 1. (76)
Again assuming that we have only a single fivebrane, the constraints for the case of

the direct sum of two line bundles — analogous to equations (56a) to (56e) — are

2

IV,= (|,|,-4)^ + >0, ^ = 1,2,3

r=l

-il + xYilWi) =0, r = l,2

+2(ai + a2) - (1 - A)Vvri) > 0,

-{l + Xya'wA >0,

a

, 2

(77a)

(77b)

(77c)

(77d)

< < a/ and <

A\2

6a^ — 6a2

1\2

5„1

a < a < 2a and

2(a2)2 -5(ai)
30ai - 12a2

< a^' <

-a

1\2

or

1^2 I ^^2X2

3aV^ + (a^

6a^ — Qa"^

(77e)

We point out that only the first two sets of equations differ from the single line bundle
case in the previous section. We must now find simultaneous solutions to equations

(77a) to (77e)

4.5 An 50(12) X U{1) X U{1) Hidden Sector

We will now focus on the specific direct sum of two line bundles given by

= L,®L2 = Ox{2, 1, 1) © OxiO, 3, 2).

In terms of our basis choice for the first Chern class, see (20), this is

(£i,£?,£?) = (2,1,1), {elelil) = (0,3,2).

(78)

(79)

21

For simplicity, we assume that there is only a single five-brane. In order to cancel the
heterotic anomaly for this choice of bundle and embedding, it must then wrap the
effective curve class

= (20,9,0) > 0. (80)

As in the previous section, we use the homogeneous rescaling (57) to set

^2/3

1.

We will also use the same five-brane position for convenience; namely A = 0.496. The
only remaining parameters are now the Kahler moduli a^, a? and a^, subject to two
Fl-term constraints and a number of inequalities. We first consider the two Fl-term

constraints (77b), which are two quadratic equations that stabilize two of the three

Kahler moduli. The standard strategy to parametrize the solution set is to pick one
variable and then compute a lexicographic Grobner basis with the chosen variable
last. If we pick as the parameter, then the result is that a? is a real solution of the
quartic equation

fa^)^ + 6.7058a^

2\3

+ 4.2352(a^)2(a2)2-0.3955(a^)^-1.5808aV-1.1801 =0 (81a)

3^2

and, finally.

1.2651(0^)^ + 8.4839a^(a2)^ + b.?,bd,2{a^fa^ - 0.83377a^ - 4a^

(81b)

It remains to impose all inequalities; namely, the positivity of all Kahler moduli.

the gauge couplings equations (77c) and (77d), as well as the visible sector stability

conditions (77e). The numerical result is that the free Kahler modulus has to lie in
the interval

< < 0.0701743. (82)

It then follows that the unique positive root a? of (81a) and a} determined by (81b)
satisfy all physical constraints. For example, if we pick = 0.06 then the remaining
Kahler moduli and gauge couplings are

/I 2 S

(a , a , a

^^(1)2' ^(2)2

(0.95,1.06,0.06),
) =(0.65,0.02).

^3)

The entire one-dimensional solution set is plotted in Figure 4

Finally, as in the one line bundle case, the particle spectrum of this low energy
5*0(10) X U{1) X U{1) hidden sector can be computed from the cohomology of the
various tensor products of Li and L2. Since this is similar to the discussion in Sub-
section 4.3, but rendered more complicated by the presence of two hne bundles, we
will not discuss the results here.

22

Figure 4' The Kdhler cone, in 3 dimensions (top) and the projection in radial directions
(bottom). The 1-dimensional blue region /C"*" is our hidden sector solution for
Li®L2 = Ox(2, 1, 1) e Ox(0,3,2) at \ = 0.496. It shows the Kdhler moduli
uj = a'^uji + a^ijJ2 + a^i^3 simultaneously satisfying the two independent FI =
conditions for the two U{1) factors, as well as the positivity of the isible and
hidden sector gauge couplings. The red region /C* is the stability region of the
visible sector bundle from Figure 1 . The intersection /C*n/C"'" is where all physical
constraints are satisfied.

23

Acknowledgments

We would like to thank Lara Anderson, Ralph Blumenhagen, James Gray, and Andre
Lukas for useful discussions. Volker Braun is supported by the Dublin Institute for
Advanced Studies and expresses his thanks to the University of Pennsylvania for its
hospitality while some this work was being carried out. Burt Ovrut is supported in
part by the DOE under contract No. DE-AC02-76-ER-03071 and the NSF under
grant No. 1001296. Yang-Hui He would like to thank the Science and Technology
Facilities Council, UK, for an Advanced Fellowship and grant ST/J00037X/1, the
Chinese Ministry of Education, for a Chang-Jiang Chair Professorship at NanKai
University, the U.S. National Science Foundation for grant CCF-1048082, as well as
City University, London and Merton College, Oxford, for their enduring support.

Appendix A Line Bundles and Eg
A.l Induced Bundles

The usual approach of constructing Eg bundles for heterotic strings is to first construct
a G-bundle for a smaller group G and then use a map (group homomorphism) ip :
G ^ Es to build an induced Es bundle. Really, this is just the composition of the
G-bundle presentation [X, BG] with Bip : BG — )■ BEg. Explicitly, we can think
of the G-bundle as a collection of transition functions ipa/3 : Uap — )■ G on overlaps
Uai3 = Ua^Up. The transition functions of the induced E^ bundle are then simply
given by the composition (pap = ipo yj^^. A popular choice for G is SU (n), equivalent
to a rank-ra vector bundle of vanishing first Chern class. This is usually combined with
a group homomorphism SU{n) C SU{9) — )■ SU{9)/'L^ C E^ that factors through the
SU (9)/Z3 subgroup corresponding to removal of a node in E^ affine Dynkin diagram.

However, there is no need to pick a special unitary group and, for the purposes
of this section, we are considering G = U{1). As a first example, the easiest group
homomorphism to E^ can be obtained from embedding U{1) as a maximal torus in
SU{2) and then embedding it further in S'f/(9) — )■ E^. Up to a choice of coordinates,
this is the homomorphism

U(l)^SU(2), e'*^(-P('^' (84)

or 1 I— 7- ( i*! ) in the Lie algebra. This construction cannot yield the most general E^
bundle, but rather only one whose structure group can be reduced to SU{9)/Z3 C £^8-
This is of course desirable for phenomenological applications, so that not all of the Eg
gauge group is broken. In this example, the commutant of SU{2) in E{8) is E-jj'Li and

^Because it will be important in this appendix, we will break with the physics tradition and be

careful about discrete quotients of groups. A good reference for the relevant group theory is 54 .

24

the embedded U{1) C SU{2) commutes with it self. Hence, the overall commutant
and unbroken gauge group of [/(I) C Es is {Ef x U{l))/Z2.

A. 2 Consistency Conditions

It is occasionally claimed that a U {n) bundle, for example, a line bundle, has to satisfy
extra divisibility conditions in order to define an induced £"8 bundle. However, this is
not true and any bundle together with a U{n) — t- Eg homomorphism is admissible. In
particular, the first Chern class of the line bundle need not be evenj^ An bundle
is automatically spin since E^ is simply connected. While it is true that a line bundle
with an odd first Chern class is not spin, any induced Eg bundle is well-defined and
admits adjoint spinorsj^ Since the U{n) bundle is purely auxiliary in this construction,
there is no need for it to be spin. This is related to the fact that the contribution
to the heterotic anomaly from a line bundle is always an even multiple of its second

Chern character, as we will discuss in ISubsection A. 3

There is a related, but different, context where an even first Chern class does play

a role 55 , 56 . In an effort to clear up any confusion, let us review it in the remainder
of this subsection. First, recall the usual conformal field theory construction of the
Eg X £"8 heterotic string. There are 36 real left-moving fermions, 16 for each Eg. The
GSO projection acts as a minus sign, so the 16 fermions transform under Spin{16) /Z2.
Only this group, and not the whole Spin{16), is a subgroup of Eg. Since it is still
difficult to construct the most general Spin{16)/Z2 bundle, it is tempting to combine
the 16 real spinors into 8 complex ones and construct a U (8) gauge bundle for them.
The trouble is that this constructs a U{S) C 50(16) gauge bundle which need not
be a Spin{l&) /'L2 bundle. A sufficient condition to avoid this problem is if the f/(8)
bundle is spin, that is, has even first Chern class. In that case it lifts to a Spin{lQ)
bundle, which we can divide to obtain a Spin{lQ)/'Z2 bundle. The key difference is
that this does not use a f/(8) — )■ Spin{16)/7j2 C Eg homomorphism to construct an
induced Eg bundle, but rather relies on a lifting that might not exist.

A. 3 Chern Classes and the Anomaly

The heterotic anomaly cancellation condition for two £'8-bundles V^^\ V^"^^ is

- C2{V^''^) - C2{V^^^) + C2{TX) = W e H\X, Z), (85)

where W is the the class of the five-brane(s). Here, C2 of a Eg bundle means its
degree-4 characteristic class. Completely analogous to the usual Chern classes, it

^That is, divisible by 2.

^We remark that this is different in the 5pm(32)/Z2 heterotic string if one uses a hne bundle
and C/(l) — ?> 50(32) homomorphism. In this case, the line bundle does have to satisfy additional
constraints for the induced bundle to lift to a 5'pm(32)/Z2 bundle.

25

is unambiguously defined in integral cohomology as the pull-back of the generator
C2 G H^{BEs,Z) ^ Z by the map [X,BEs] defining the bundle. Fortunately, we
never have to actually evaluate the homotopy-theoretic definition. For an Eg bundle
whose structure group reduces to the S'[/(9)/Z3 subgroup, which is the case we are
interested in for phenomenolgical reasons, the degree-4 characteristic class of the Eg
bundle coincides with the usual second Chern class of the SU{9) bundle. Hence, the
contribution to the anomaly of such an induced bundle Vp defined by a U{n) bundle
V and homeomorphism p : U{n) — )■ SU{'9) can be computed in terms of the Chern
classes of V and the group theory of p.

Let us consider the case where V^*^^-* = Lp is induced from a line bundle L and
p : U{1) — )■ SU{9). On general grounds, the anomaly cancellation condition then
must be of the form

ap ciiiy - C2(\/(2)) + ^^^jx) = W e H\X, Z) (86)

since the only available characteristic class is Ci (L) with some group-theoretic numeri-
cal coefficient ap. In de Rham cohomology the visible sector contribution is represented
by

a^Ci(L)2 = 2ap ch2(L) = -^tr.FAF (87)

167r^

It is suggestive, but wrong, that the coefficient Op should be half-integral such that
the contribution of the line bundle to the anomaly is always an integer multiple of
its second Chern character. In fact, the coefficient Op is always integer which is twice
what one would naively expect]^ This is also required for the anomaly contribution
apCi(L)^ to define an integral cohomology class. As the simplest example, let us return

to the p : f/(l) ^ SU{2) C SU{9) embedding from (|84|). The induced SU{9) bundle
is

Lp = L®L-^® I7, (88)
so its second Chern class is —C2{Lp) = cf(L), that is, Op = 1.

A. 4 Example

As a more complicated example, we now consider a combination of line bundle and
non-Abelian bundle. Let us start with a SU{3) bundle V and a line bundle L. We
can use this data with different group homomorphisms to construct the same SU{9)
(and, therefore. Eg) bundle in two different ways:

(A) Use the group homomorphism

Pa : SU{3) x f/(l) ^ SU{A) C SU{9),

(^73x3,e'*) H> I

15x5

^9)

^For discrete Wilson lines, this observation was already made in the footnote on Page 88 of [st] .

26

(B) The direct sum {V ^L)(BL ^ is a rank-4 bundle with vanishing first Chern class.
We combine it with the trivial embedding pB : S'f/(4) C SU{9).

Both constructions yield the same induced SU (9)-bundle, namely

E = {V^L)®L-^®l5 = {V® L)p^ = {{V ® L) © L-^)^^. (90)

Its contribution to the heterotic anomaly is —C2{E) = —C2(y) + 3ci(L)^. In other
words, the group-theoretic coefficient Op = 3. It is again an integral class, as it must
be.

Bibliography

[1] D. J. Gross, J. A. Harvey, E. J. Martinec, and R. Rohm, "Heterotic String
Theory. 1. The Free Heterotic String," Nucl. Phys. B256 (1985) 253. |l]

[2] D. J. Gross, J. A. Harvey, E. J. Martinec, and R. Rohm, "Heterotic String
Theory. 2. The Interacting Heterotic String," Nucl. Phys. B267 (1986) 75. [l|

[3] P. Horava and E. Witten, "Heterotic and type I string dynamics from eleven
dimensions," Nucl. Phys. B460 (1996) 506-524, hep-th/9510209, [l|

[4] P. Horava and E. Witten, "Eleven-dimensional supergravity on a manifold with
boundary," Nucl.Phys. B475 (1996) 94-114, hep-th/9603142. [l|

[5] A. Lukas, B. A. Ovrut, and D. Waldram, "On the four-dimensional effective
action of strongly coupled heterotic string theory," Nucl. Phys. B532 (1998)
43-82, .hep-th/9710208, [l]

[6] A. Lukas, B. A. Ovrut, and D. Waldram, "The ten-dimensional effective action
of strongly coupled heterotic string theory," Nucl. Phys. B540 (1999) 230-246,
fhep- th/98010 87, [l]

[7] A. Lukas, B. A. Ovrut, and D. Waldram, "Nonstandard embedding and
five-branes in heterotic M theory," Phys. Rev. D59 (1999) 106005,
|hep-th/9808101, [T|[3J]

[8] A. Lukas, B. A. Ovrut, K. Stelle, and D. Waldram, "Heterotic M theory in
five-dimensions," Nucl.Phys. B552 (1999) 246-290, hep-th/980605 1. [II [31]

[9] A. Lukas, B. A. Ovrut, K. Stelle, and D. Waldram, "The Universe as a domain
wall," Phys. Rev. D59 (1999) 086001, hep-th/9803235, [31]

[10] S. K. Donaldson, "Anti self-dual Yang-Mills connections over complex algebraic
surfaces and stable vector bundles," Proc. London Math. Soc. (3) 50 (1985),
no. 1, 1-26. [1]

27

[11] K. Uhlenbeck and S.-T. Yau, "On the existence of hermitian- Yang-Mills
connections in stable vector bundles," Comm. Pure Appl. Math. 39 (1986),
no. S, suppl., S257-S293. Frontiers of the mathematical sciences: 1985 (New
York, 1985). |l|

E. Witten, "Strong coupling expansion of Calabi-Yau compactification,"
Nucl.Phys. B471 (1996) 135-158, hiFth7960 2070[ [l|

B. R. Greene, K. H. Kirklin, P. J. Miron, and G. G. Ross, "A Superstring
Inspired Standard Model," Phys. Lett. B180 (1986) 69. [l|

B. R. Greene, K. H. Kirklin, P. J. Miron, and G. G. Ross, "A Three Generation
Superstring Model. 1. Compactification And Discrete Symmetries," Nucl. Phys.
B278 (1986) 667. [l|

B. R. Greene, K. H. Kirklin, P. J. Miron, and G. G. Ross, "A Three Generation
Superstring Model. 2. Symmetry Breaking And The Low-Energy Theory,"
Nucl. Phys. B292 (1987) 606. [f|

T. Matsuoka and D. Suematsu, "Realistic Models From The x E'^
Superstring Theory," Prog. Theor. Phys. 76 (1986) 886. [l]

B. R. Greene, K. H. Kirkhn, P. J. Miron, and G. G. Ross, "27^ Yukawa
Couplings For a Three Generation Superstring Model," Phys. Lett. B192
(1987) 111. |l|

V. Braun, Y.-H. He, B. A. Ovrut, and T. Pantev, "The Exact MSSM spectrum
from string theory," JHEP 0605 (2006) 043, hep-th/0512177, [l| g

V. Braun, Y.-H. He, and B. A. Ovrut, "Stability of the minimal heterotic
standard model bundle," JHEP 0606 (2006) 032, hep-th/0602073, [T| [2| [3^

V. Braun, Y.-H. He, and B. A. Ovrut, "Yukawa couplings in heterotic standard
models," JHEP 0604 (2006) 019, hep -th/0601204| [l|

J. Gray, A. Lukas, and B. Ovrut, "Perturbative anti-brane potentials in
heterotic M-theory," Phys. Rev. D76 (2007) 066007, [hipth/07 01025"', |l|

J. Gray, A. Lukas, and B. Ovrut, "Flux, gaugino condensation and anti-branes
in heterotic M-theory," Phys. Rev. D76 (2007) 126012, [ 0709 . 29 14[ [l|

V. Braun and B. A. Ovrut, "Stabilizing moduli with a positive cosmological
constant in heterotic M-theory," JHEP 0607 (2006) 035, |hep-th/06030 88, |l|

28

S. Kachru, R. Kallosh, A. D. Linde, and S. P. Trivedi, "De Sitter vacua in
string theory," Phys.Rev. D68 (2003) 046005, hep-th/030124 0, |l|

F. A. Bogomolov, "Holomorphic tensors and vector bundles on projective
manifolds," Izv. Akad. Nauk SSSR Ser. Mat. 42 (1978), no. 6, 1227-1287, 1439.
□

M. R. Douglas, R. Reinbacher, and S.-T. Yau, "Branes, bundles and attractors:
Bogomolov and beyond," ,math/0604597, [l]

B. Andreas and G. Curio, "Spectral Bundles and the DRY-Conjecture,"
J.Geom.Phys. 62 (2012) 800-803, 1012.3858, |l|

B. Andreas and G. Curio, "On the Existence of Stable bundles with prescribed
Chern classes on Calabi-Yau threefolds," |1104.34 35. [l]

R. Blumenhagen, G. Honecker, and T. Weigand, "Loop-corrected
compactifications of the heterotic string with line bundles," JHEP 0506 (2005)
020, hep-th/0504232t [l| |3]3} [3^ [3^ [3^

R. Blumenhagen, G. Honecker, and T. Weigand, "Non-Abelian brane worlds:
The Heterotic string story," JHEP 0510 (2005) 086, hep -th/05 10049, |l| ^

R. Blumenhagen, S. Moster, and T. Weigand, "Heterotic GUT and standard
model vacua from simply connected Calabi-Yau manifolds," Nud.Phys. B751
(2006) 186-221, [hi^-th/0603015j |l| [S^

R. Blumenhagen, S. Moster, R. Reinbacher, and T. Weigand, "Massless Spectra
of Three Generation U(N) Heterotic String Vacua," JHEP 0705 (2007) 041,
[hip^th/0612039( [l| |3]^ [3^

T. Weigand, "Compactifications of the heterotic string with unitary bundles,"
Fortsch.Phys. 54 (2006) 963-1077. [I| [3l|

L. B. Anderson, J. Gray, A. Lukas, and B. Ovrut, "Stability Walls in Heterotic
Theories," JHEP 0909 (2009) 026, |0 905 . 174 8. |l|

L. B. Anderson, J. Gray, and B. Ovrut, "Yukawa Textures From Heterotic
Stability Walls," JHEP 1005 (2010) 086, |1001. 23 171 [l|

L. B. Anderson, J. Gray, A. Lukas, and E. Palti, "Two Hundred Heterotic
Standard Models on Smooth Calabi-Yau Threefolds," Phys.Rev. D84 (2011)
106005, 11106.48041 [1]

L. B. Anderson, J. Gray, A. Lukas, and E. Palti, "Heterotic Line Bundle
Standard Models," JHEP 1206 (2012) 113, .1202 . 1757, [l|

29

[38] L. B. Anderson, J. Gray, A. Lukas, and E. Palti, "Heterotic standard models
from smooth Calabi-Yau three-folds," PoS CORFU2011 (2011) 096. [l]

[39] R. Donagi, A. Lukas, B. A. Ovrut, and D. Waldram, "Nonperturbative vacua
and particle physics in M theory," JHEP 9905 (1999) 018, hep-th/9811168, [l|

[40] R. Donagi, A. Lukas, B. A. Ovrut, and D. Waldram, "Holomorphic vector
bundles and nonperturbative vacua in M theory," JHEP 9906 (1999) 034,
[hep-th/9901009, [l]

[41] R. Donagi, B. A. Ovrut, T. Pantev, and D. Waldram, "Standard models from
heterotic M theory," Adv.Theor.Math.Phys. 5 (2002) 93-137, ,hep-th/9912208(

m

[42] R. Donagi, B. A. Ovrut, T. Pantev, and R. Reinbacher, "SU(4) instantons on
Calabi-Yau threefolds with Z(2) x Z(2) fundamental group," JHEP 0401 (2004)
022, hep-th/0307273 . [l]

[43] V. Braun, B. A. Ovrut, T. Pantev, and R. Reinbacher, "Elliptic Calabi-Yau
threefolds with Z(3) x Z(3) Wilson lines," JHEP 0412 (2004) 062,
rhep-th /0410055 . [T[ |2T

[44] R. Donagi, Y.-H. He, B. A. Ovrut, and R. Reinbacher, "The Spectra of

heterotic standard model vacua," JHEP 0506 (2005) 070, hep-th/0411156j [l|

[45] C. Schoen, "On fiber products of rational elliptic surfaces with section," Math.
Z. 197 (1988), no. 2, 177-199.

2.1

[46] B. A. Ovrut, T. Pantev, and R. Reinbacher, "Torus fibered Calabi-Yau
threefolds with nontrivial fundamental group," JHEP 0305 (2003) 040,
[hep-th/0212221, [2l]

[47] V. Braun, M. Kreuzer, B. A. Ovrut, and E. Scheidegger, "Worldsheet

Instantons, Torsion Curves, and Non-Perturbative Superpotentials," Phys. Lett.

B649 (2007) 334-341, ,hep-th/0703134, 2.1

[48] V. Braun, M. Kreuzer, B. A. Ovrut, and E. Scheidegger, "Worldsheet

instantons and torsion curves. Part A: Direct computation," JHEP 10 (2007)

022, hep-th/0703182, 2.1

[49] V. Braun, M. Kreuzer, B. A. Ovrut, and E. Scheidegger, "Worldsheet

Instantons and Torsion Curves, Part B: Mirror Symmetry," JHEP 10 (2007)

023, arXiv : 0704 . 0449 [hep-th] , [21

30

[50] T. L. Gomez, S. Lukic, and I. Sols, "Constraining the Kahler moduli in the
heterotic standard model," Commun.Math.Phys. 276 (2007) 1-21,
|hep-th/05 12205, |2T

[51] M. B. Green, J. H. Schwarz, and E. Witten, "Superstring Theory. Vol. 1:
Introduction,". Cambridge, Uk: Univ. Pr. ( 1987) 469 P. ( Cambridge

Monographs On Mathematical Physics). |2.2, 3.1

[52] M. B. Green, J. H. Schwarz, and E. Witten, "Superstring Theory. Vol. 2: Loop
Amplitudes, Anomalies And Phenomenology,". Cambridge, Uk: Univ. Pr. (

1987) 596 P. ( Cambridge Monographs On Mathematical Physics). 2.2, 3.1

[53] R. Blumenhagen, G. Honecker, and T. Weigand, "Supersymmetric

(non-)Abelian bundles in the Type I and SO (32) heterotic string," JHEP 0508

(2005) 009, hep-th/0507041 3.3

[54] J. Distler and E. Sharpe, "Heterotic compactifications with principal bundles
for general groups and general levels," Adv. Theor. Math. Phys. 14 (2010)
335-398, |hep-th/0701244l [2]

[55] J. Distler and B. R. Greene, "Aspects of (2,0) String Compactifications,"

Nucl.Phys. B304 (1988) 1. A.2

[56] D. Freed, "Determinants, Torsion, and Strings," Commun.Math.Phys. 107

(1986) 483-513. A.2

[57] E. Witten, "Global Anomalies in String Theory," . [s]

31